Carlos L’Abbate
Here’s a frank assertion: The United States cannot reasonably hope to meet its decarbonization goals unless its electric utilities deploy modern, private, wireless broadband communications networks. Drawing that connection between decarbonization and broadband capabilities is less complicated than it sounds (Figure 1). In short, addressing climate change requires reduced reliance on fossil fuels to generate electricity and more reliance upon renewable resources, which the grid is not designed to fully accommodate. Ramping up renewables requires a smarter grid and that means using millions of widely distributed sensors to create data, applications to analyze data, and smart devices to take action based upon data. Collecting and communicating all that data requires—here it is—a utility-grade wireless broadband network.
Regulatory mandates and utilities’ voluntary commitments include major decarbonization progress in the relatively near term, with milestones at 2030, 2035, and 2050. Considering that it took commercial cellular companies many years to build out coverage nationwide, time is of the essence for utilities to deploy broadband networks in their service territories that meet their own critical infrastructure requirements. This article explains how we can get there.
The chain of causation is direct: in response to climate change, utilities are moving to reduce their emissions of greenhouse gases, including carbon dioxide and methane. Decarbonization requires reduced dependence upon fossil-fuels to generate electricity and vastly greater deployment of cleaner, renewable resources, such as solar photovoltaic and wind power. These renewables are typically smaller and far more distributed than the traditional large, central power plant, and they also can be intermittent. Thus, where the utility historically sent centrally produced power in one direction over the grid to the end-user, it is now set to become the master orchestrator of a multitude of distributed energy resources (DERs). These DER “prosumers” can morph from producers to consumers of electricity as quickly as a cloud can block the sun. The nation’s grid, first built a century ago, was not designed with this new role in mind. Figure 2 illustrates this very significant change.
Today, a utility must safely and efficiently support the two-way flow of electricity over the grid. Operators thus require greatly enhanced grid visibility, control, and automation capabilities. The sensors, smart devices, and applications that will provide operators these enhanced capabilities depend upon connectivity via a data network. Such sensors and devices are already proliferating well beyond those deployed even a few years ago: a one-million meter investor-owned utility will see an eightfold increase in connected endpoints between 2018 and 2028. Increasing sophistication and number of applications and endpoints means vastly more data traversing the network, and increased automation requires that the data be communicated in real time with extremely low latency.
Electric utilities across the country have long operated private wireless networks to support grid operations. These legacy, narrowband (low-capacity) networks are not designed to provide the low latency and capacity required to support the flow of data to enable the modern grid. Rather, as Figure 3 illustrates, they are generally intended to support only the application for which they were initially deployed and cannot accommodate additional, data-intensive, and often mission-critical applications with rigorous latency requirements, nor can they scale or help future-proof the utility. These networks frequently are proprietary, use unlicensed or very narrow swaths of spectrum, and are at end-of-life, and thus are subject to reduced support or even vendor obsolescence.
Modern wireless broadband services are available from commercial carriers, but for reasons unrelated to the technology, they too are unable to meet the requirements of the modern grid. Commercial networks may provide service with low latency and adequate capacity, but they cannot provide the level of control utilities required to ensure network availability and security. Commercial networks are designed to meet the needs of the consumer mass market, not the more rigorous requirements of the nation’s most critical infrastructure industry. For example, commercial networks may not provide coverage in sparsely populated areas, despite the presence of utility infrastructure that requires connectivity. Similarly, commercial carriers typically deploy only those security measures that do not overly inconvenience their mass market subscribers; utilities require the ability to implement more rigorous—and frequently more burdensome—protections. Finally, commercial carrier services sunset: 2G gave way to 3G, and 3G is the latest to be discontinued. When these network evolutions occur, they are on the carrier’s schedule, forcing customers to replace the old devices, often at substantial cost.
And then there is cost and efficiency. Because legacy networks tend to be single-purpose, most utilities maintain many of them; one utility, for example, was operating more than 20 separate legacy systems in 2021. Even without a grid modernization imperative, cost and inefficiency will make maintaining multiple legacy networks unsustainable, particularly in light of the forecasted sharp growth in connectivity needs.
As the above paragraphs suggest, the grid communications solution utilities require must be broadband (high capacity) and it must be private. And where cost, terrain, or device density makes fiber connectivity unattractive, private LTE wireless broadband can meet the requirements of the modern grid.
LTE is a global standard for cellular communications. A proven, mature technology deployed by carriers worldwide, LTE benefits from a broad ecosystem of vendors, goods, and services (Figure 4).
A private LTE network is owned, operated, and entirely controlled by the utility for its own grid management purposes. In practice, the network will help enable the utility to capture, aggregate, analyze, and act upon data created by sensors installed throughout the grid using the LTE communications standard. Leveraging data analytics, artificial intelligence, and machine learning to model and exercise the transmission and distribution systems, the utility will be better able to forecast failures, prevent outages, and minimize truck rolls. Private LTE will allow a utility to understand the impact of distributed energy generation resources, even when not owned by the utility. The utility will be able to obtain true visibility into and control over the grid, all because the private, high-performance LTE network can carry that data to its applications in a secure and reliable fashion with latency low enough to support the most time-sensitive mission critical use cases.
Spectrum is a collection of radio frequencies that are the indispensable foundation upon which all wireless networks operate. A private LTE network cannot work without spectrum, and the specific spectrum the utility selects will impact the design and performance of the network, as well as the business model for sustaining it. And—importantly—it will determine the extent to which the utility can control and benefit from its network (Figure 5).
The Federal Communications Commission (FCC) regulates spectrum use, and it does not treat all bands the same. Most utilities will insist on licensed spectrum, which generally provides an exclusive right to use the spectrum with legal recourse through the FCC if an unauthorized system encroaches on the utility’s use. Unlicensed spectrum, by contrast, is open to the public, and systems that use unlicensed spectrum (such as consumer Wi-Fi, Bluetooth, and cordless telephone systems) are widely available in the consumer market. As a result, the use of unlicensed spectrum is typically subject to contention from other networks, depriving the utility network of guaranteed full and unimpeded access to the specified frequencies.
This regulatory benefit is why utilities generally eschew unlicensed spectrum for their mission-critical communications, like those controlling the grid: they simply cannot allow potential congestion or interference to prevent or delay the transmission and receipt of data. If a utility does not control the spectrum, it cannot control the network. Licensed spectrum provides that control.
As suggested above, a private LTE network enables the utility to ensure coverage where it is needed. As a matter of both the network’s geographic reach and the number of devices it can accommodate at any particular location, a utility’s determinative coverage factors may differ substantially from those of a commercial LTE service provider. A utility must ensure connectivity to devices installed on the grid, often in remote locations; a commercial carrier will focus on connectivity where the population is sufficient to support the expense of deploying and maintaining the coverage. A private network allows the utility to design and build coverage to meet its own requirements, without consideration of the business needs of a commercial carrier.
A private LTE network can also provide the network performance the utility needs. It can provide the bandwidth and the latency required by the applications the utility will run across the network. San Diego Gas & Electric, for example, has announced its intent to deploy a new application on its private LTE network. It will deploy falling conductor protection, which requires low-latency broadband connectivity to de-energize a falling line before it hits the ground and can cause a wildfire. According to Schweitzer Engineering Laboratories, which created the application, the technology has long existed, but it wasn’t until private LTE—and licensed spectrum to enable it—was available to utilities that there was a workable wireless connectivity option with sufficiently low latency to support it.
Another aspect of performance is the ability to prioritize traffic over the network so that communications of the greatest criticality are first in line for network resources. Because the network is private, the utility can control the network’s configuration to prioritize those communications that it—and only it—considers most important.
Cyberattacks are evolving and becoming increasingly sophisticated. For legacy networks, this situation can be a significant problem; as vendors reduce and discontinue support for older communications systems, keeping them patched and secure from new exploits becomes difficult. Migrating to a state-of-the-art private LTE network hardened to utility requirements could significantly improve a utility’s security posture.
LTE is the first cellular standard developed for cybersecurity at the outset, instead of added on at the end of the process. In addition, the LTE’s international partnership of standards bodies constantly reviews and revises LTE to address new and emerging cyber threats.
LTE offers state-of-the-art security capabilities, some of which are optional within the LTE standard and not necessarily deployed by commercial carriers. However, when the utility owns, operates, and controls its own private LTE network, it can choose to adopt cyber protections as stringent as it desires to meet its critical infrastructure function. Because these optional features can be burdensome on users and/or operators, deciding whether to implement them requires some balancing of that burden against the security benefit. Thus, while a commercial carrier serving the mass market may not be willing to impose such a burden on its consumer subscribers, an electric utility will have a different calculus. Private LTE allows a utility, at its own discretion, to impose even the strongest and most burdensome protections without considering the preferences of any other entity.
A utility adopting a private network can also implement one particularly valuable protection that has nothing to do with LTE. As a quick review of the headlines suggests, many attacks on critical infrastructure systems are launched remotely over the Internet. A longstanding best practice in the utility industry is to isolate networks that are used for mission-critical grid operations from the Internet. Known as air-gapping, this measure can be implemented only in a private-network scenario; commercial networks are purpose-built for Internet access.
In addition to cybersecurity, a utility must also consider the physical security of its network. Even with an air-gapped network, cell sites require fencing, surveillance, and other measures to protect against physical attacks. Utilities have long been the nation’s model for the physical security of critical assets; because the private LTE network will be such a critical asset, a utility can apply its well-developed physical security standards and practices to elements of the network.
The growing interest among utilities in private LTE raises the possibility that coordinated, broad adoption of private LTE networks could unleash a new category of benefits made possible by enabling utility communications networks to interoperate. Four major investor-owned utilities (Ameren, Evergy, San Diego Gas & Electric, and Xcel Energy) have executed agreements for the use of licensed 900-MHz spectrum for their private LTE networks, and 12 additional entities have or have had experimental licenses to test 900-MHz private LTE. Where utility networks use the same technology and band, they can “talk” to each other, opening an array of possibilities, ranging from the current and known to the futuristic and visionary (Figure 6).
Some benefits of such a “network of networks” are easily understood; others may not become clear until well into the future. In the short term, for example, repair crews from a neighboring utility providing mutual aid after a significant storm could, with appropriate authorization, use their own mobile devices to access the host utility’s critical operational systems (“roaming” from one network to another). Such access could even provide a “mobile desktop” experience, greatly increasing efficiency and reducing recovery time. (Interestingly, the service territories of Ameren and Evergy—both of which have moved to deploy 900-MHz private LTE networks—border each other, perhaps creating an opportunity for an early network of networks investigation.) With broad adoption of private LTE, utilities also could share use of specific, cloud-based network elements, rather than each having to own, operate, and maintain every element itself. In the longer term, a utility may be able to securely communicate with out-of-territory electric vehicles as they traverse its service area, helping predict and manage charging loads and vehicle-to-grid power supply.
In addition to interoperability, a utility will find other benefits in utilizing a band and technology that is broadly adopted across the industry, including wide choice and vibrant innovation in the marketplace for applications and equipment that work on their networks. The resulting ecosystem creates a virtuous cycle: more utilities adopting private LTE bring more vendors to the marketplace, and the choice and innovation provided by a more robust marketplace bring more utilities to private LTE. And finally, whether utilities are focused on the macro issues of resiliency, cybersecurity, and decarbonization or the sharing of infrastructure, talent, knowledge, and technology, a network of individual networks can drive enhanced innovation and value to each participating utility.
A range of initiatives would help foster growth of such a network of networks, including the following:
The above discussion explains how a private LTE network can help a utility meet its communications needs. This section focuses on factors that will play into the design and construction of the network. The short answer is this: the design is determined by the purposes for which the utility is building the network in the first place.
The most important considerations in private LTE network design will be the use cases: the specific tasks the utility wants the network to accomplish and the applications it wants the network to support. What network capabilities do those use cases require? The use cases will determine network design.
A utility will likely consider many private LTE use cases with requirements ranging from “basic” to “complex.” In general, those use cases that involve mobility, mission criticality, and capacity (large volumes of data) require more complex networks. Such complexity will affect choices related to a range of elements, including at a very high level, the following:
As a general rule, the use case with the most complex requirements will drive the network design throughout. A straightforward example is: a network sufficient to support a residential meter application might provide fair coverage in a residential area but none out on the interstate highway; a network built to support a mission-critical mobile application, in contrast, will require strong coverage both on the highway as well as in the residential area. In short, a utility will typically design the network for its most challenging use case, providing communications capabilities more than sufficient for the less-demanding use cases. Figure 7 names some common use cases in terms of the complexity of network each requires.
As described previously, in order to be “private” and provide the control and capabilities utilities typically require, an LTE network must use licensed, dedicated spectrum. But as with other network design questions, the intended network uses and required capabilities will guide the choice among licensed spectrum options.
When a utility seeks spectrum for its private LTE network, it is trying to identify a contiguous block of enough radio frequencies to support a private LTE deployment of sufficient bandwidth to meet the utility’s data communications needs. Spectrum is identified by the frequency of its radio waves as measured in hertz; the amount of spectrum in a block is measured by subtracting the frequency at the bottom of the block from the frequency at the top of the block, also measured in hertz. Thus, a 500-MHz block of spectrum might extend from the 2.3-GHz frequency to the 2.8-GHz frequency. For broadband, the LTE standard currently requires at least 1.4-MHz of spectrum in two separate blocks (one for transmitting and one for receiving); in the 900-MHz band; for example, 6 MHz of broadband spectrum are available in two blocks of 3 MHz each at 897.5–900.5 MHz and 936.5–939.5 MHz.
The critical point to understand here is that spectrum bands differ, as a matter of both physics and regulation. A utility must choose spectrum carefully because it will serve as the foundation supporting the new private LTE network and will greatly impact the network design necessary to meet the utility’s communications requirements. Because of the nature of radio waves, signals in low-band spectrum (below 1 GHz) can travel farther than signals in higher-band spectrum and can better penetrate obstacles (such as buildings and foliage). As a result of these “propagation characteristics,” networks that use low-band spectrum require fewer cell sites (towers with antennae) than do networks using higher-band spectrum, thus significantly reducing the system’s total cost of ownership.
Regulation of spectrum bands varies as well. The FCC organizes a band into wider or narrower channels, which can determine whether it is appropriate for broadband services. To facilitate and coordinate spectrum usage in a way that controls radio interference among systems, the FCC also sets technical limits on the systems that use specific bands, including restrictions on the power of signals transmitted over the band. These rules can affect both the cost and performance of the utility network.
Many utilities will have specific areas where many devices will need to access the network. Such high-density locations might include cities and generation facilities, for example, where there might be a greater need for large numbers of connected sensors. Much of the utility’s service territory, however, may be sparsely populated, with grid facilities that are more distributed and a lower density of connected devices. No matter where the device, sensor, or application is located (and for some applications the device might be in motion), if it will need connectivity, the network must be designed to provide coverage for it, whether the application exists today or is planned to exist in the future (Figure 8).
Coverage is not a binary feature, however. As anybody with a cell phone knows, coverage at a certain location may be excellent, nonexistent, or any degree in between. Some applications may not function properly if an associated device has less than optimal connectivity; others may be more forgiving. Because coverage comes with a cost—the investment in building or leasing a cell tower, for example, and the expense of maintenance—network designers working within a budget will need to make tradeoffs, strategically placing infrastructure where it can best meet the coverage needs of the utility’s primary use cases.
Ultimately, the purpose of the utility private LTE network is to enable wireless communications to and from devices in the field, whether they be sensors monitoring voltage, circuit status, or weather conditions that only provide data, or active devices (like circuit reclosers) that take a physical action on the grid based upon received data. The devices must include a module designed to transmit and receive LTE communications over the selected spectrum to connect over the network. That module will include a microchip containing information about the device’s identity so that the network can determine how to treat the device’s request to connect and its payload traffic: for example, the priority level to assign to data from this device. In building and operating its private LTE network, a utility will need to focus on the selection of devices, their integration to the network and supported applications, and their ongoing management.
A utility may be connecting thousands or tens of thousands of devices to its new private LTE network, so keeping track and managing them can be a challenge. A device management platform is software and firmware that handles tasks like fulfilling, provisioning, activating, and patching devices, all over the air via the private LTE network. Many such platforms are available in the marketplace and are typically acquired separately from the LTE radio access network and core.
Embedding LTE modules into devices that the network will support requires effort and investment from application vendors, including the cost of obtaining federal regulatory approval to sell the device for use in a particular spectrum band. In some cases, the module may be external but connected to the device, but most often—and most efficiently and securely—the module will be embedded within the device. Because the device will be specific to the application, this integration will be accomplished by the device/application vendor.
To understand the importance of efficient integration, consider again the example of falling conductor protection, the wildfire mitigation application from Schweitzer Engineering Laboratories that senses a falling line and de-energizes it before it reaches the ground. The application requires a remote protective relay to collect phasor data for each line segment and send it to the centrally located controller running the falling conductor detection algorithm, which then returns a signal to a remote switch to de-energize the line. Those signals are wireless communications enabled by integrated 900-MHz LTE modules embedded within the protective relay, the controller, and the switch. Since the entire process must be completed before the line hits the ground, the communication must be extremely low latency, so efficient integration is critical.
Sometimes a use case arises from a business model, rather than a desired grid management function. Electric utilities have long been interested in the idea of using their private communications networks to support other nonelectric utilities within their service areas. In the past, the technology focus was on mesh networks; today, with private LTE, there is a standards-based, scalable way to make this vision a reality. So, for example, consider a gas or water utility serving an area overlapping with an electric utility’s service territory. The electric utility might allow the gas or water utility to use the private LTE network to support its own operations, whether mission-critical or perhaps less-critical functions like meter reading (Figure 9).
The electric utility could even use its private LTE network to provide billing services for the gas or water utility, not just connectivity. Such external uses—which could create new revenue streams—could affect network requirements like capacity and coverage, so they should be considered in network design.
Building out the full extent of planned coverage for a private LTE network will take time. Nonetheless, a utility may have a pressing need to deploy and activate a new application in a particular area, a requirement that cannot wait until private LTE coverage is activated there. In the meantime, operating an application over a commercial cellular network may be an acceptable short-term solution.
Devices identify themselves and assert permission to connect to a network through a code (called an international mobile subscriber identity or IMSI number) contained on a subscriber identity module (SIM) card. A device “homed” on a utility’s private LTE network will have a SIM card with the IMSI number corresponding to that network. With the carrier’s permission, the utility’s devices can connect to the carrier’s network if either: 1) the devices’ SIMs also include the carrier network’s IMSI or 2) the carrier allows devices with your network IMSI to “roam” onto the carrier network.
Dispatching a crew to a remote location on the grid for the initial deployment of a device (using a carrier network) can be costly. To avoid the need for a second truck roll to switch that device to the utility’s private LTE network when its coverage in the area is activated, a utility may wish to employ one of the above options. Because utilities frequently do not possess in-house experience or expertise in the carrier negotiations required to set up these arrangements, they may wish to procure third-party solutions that address this issue, which arises as they begin to transition to new private LTE networks.
With the clock ticking, utilities have only a few years to meet their first set of decarbonization targets. The required transition to renewable energy resources can only happen if the grid gets smarter, and those “smarts” will depend upon reliable, secure, modern wireless broadband networks. As a result, utilities have no time to waste in getting their network planning, design, and deployment efforts.
Private LTE offers a ready, standards-based solution: a mature, proven technology supported by a vast global ecosystem of goods, services, and expertise. Utilities should evaluate their use cases, select the spectrum band they wish to adopt, and move promptly to design and implement their networks. The nation’s ability to meet its climate change goals depends upon utilities’ rapid deployment of smart grid capabilities. And the modern grid requires modern communications.
“Operating a private LTE network,” Utility Broadband Alliance, Durham, NC, USA, 2022. [Online.] Available: https://www.ubba.com/wp-content/uploads/2022/09/Operating-PLTE-Network-paper-final-AUG-2022.pdf
“Communications requirements of smart grid technologies,” U.S. Department of Energy, Washington, DC, USA, Oct. 2010. [Online.] Available: https://www.energy.gov/sites/default/files/gcprod/documents/Smart_Grid_Communications_Requirements_Report_10-05-2010.pdf
Navigant, Utility C-Suite Take Note, “Private LTE wireless networks are required to address cybersecurity threats, grid modernization and national disasters,” Anterix, Woodland Park, NJ, USA, 2019. [Online.] Available: https://anterix.com/navigant-research-p-lte/
Carlos L’Abbate is with Anterix, Woodland Park, NJ 07424 USA.
Digital Object Identifier 10.1109/MPE.2023.3288589
Date of current version: 21 August 2023
1540-7977/23©2023IEEE